JP2018109898A - Data migration system - Google Patents

Abstract

The present invention provides a systematic and step-by-step database migration when migrating from a mainframe computer system to an open system, which can reduce the man-hours for migration and perform step-by-step and reliable migration. A data migration system includes means for reading a data structure and data of a structural database, and means for converting the data structure and data of the structural database into a data structure and data of a relational database equivalent thereto. . For this purpose, there is means for associating and holding records and data items in the structural database and tables and fields in the relational database. In addition, there is a means for obtaining an update of a structural database from a certain point in time and converting the structural database of the updated database into data and data of a relational database equivalent to the data. . [Selection] Figure 8

Description

  The present invention relates to a data migration system, and more particularly to a data migration system suitable for use in migrating a database in an open system environment from a database in a mainframe system environment.

  With the dramatic change in IT environment in recent years, computer systems at the core of enterprises are rapidly shifting from mainframe computer systems to so-called open systems. Under such circumstances, many companies have recently promoted migration from mainframe system environments to open system environments such as Windows (registered trademark) and Linux (registered trademark), which are mainly open systems. As a migration method, there is a method of collectively migrating an application system from a mainframe environment to an open environment, but this is difficult in an environment where a plurality of applications are operated. Therefore, it is also effective to perform migration in stages, and keeping the configuration of the heterogeneous system in the same state when performing migration in stages is very effective in improving the quality of migration.

  For example, in the migration system described in Patent Document 1, when transferring file data in heterogeneous migration, the consistency check of the target file and the file item definition document and the correction of the error part are performed, and the man-hours related to the migration are reduced. We provide technology that can reduce quality and prevent quality degradation that arises from defects in the configuration of heterogeneous systems.

JP 2014-225171 A

  When migrating from mainframe computer systems to open systems, the most important issue is database migration. In particular, a mainframe computer system uses a structured database, and an open system migrates to a general relational database (relational database (RDB)). Because the models are different, the migration effort is significant. Conventionally, such a systematic migration system has not been considered for migration from a structured database to an open system.

  In addition, when performing long-term migration, the migration source and migration destination systems are operating in parallel, resulting in configuration differences between different data systems, data synchronization failures and migration destinations. The system to be built may be built based on an old configuration, leading to degradation of migration quality.

  In the prior art described in Patent Document 1, migration is simplified by migrating file data, but it is impossible to prevent the occurrence of wrinkles in the database configuration.

  The object of the present invention is to perform database migration systematically and step by step when migrating from a mainframe computer system to an open system, reducing the number of migration steps, and performing data migration that enables stepwise and reliable migration To provide a system.

  The configuration of the data migration system of the present invention comprises means for reading the data structure and data of the structural database, and means for converting the data structure and data of the structural database into the data structure and data of the relational database equivalent thereto. It is what you have.

  Preferably, the data migration system includes means for associating and holding records and data items of the structural database and tables and fields of the relational database.

  Preferably, in the data migration system, an update from a certain point in time to the structural database is acquired, and the data structure and data of the structural database for the update are compared with the equivalent data structure of the relational database. Means for converting to data.

  According to the present invention, data migration can be performed in a systematic and step-by-step manner when migrating a database from a mainframe computer system to an open system, thereby reducing the number of migration steps and performing step-by-step and reliable migration. A system can be provided.

It is a figure showing the whole data migration system lineblock diagram. It is a hardware / software block diagram of a structure information management conversion server. It is a figure explaining contrast of a structure type database and a relational type database. It is a figure which shows the example of a mainframe database structure information table. It is a figure which shows the example of a mainframe record structure information table. It is a figure which shows the example of a table definition link information table. It is a figure which shows the example of an update part table definition link information table. It is a flowchart which shows the process at the time of transfer of a database from a structural type database to a relational type database. It is a flowchart which shows a process when the structure database is updated or the setting regarding transfer is changed. It is a figure which shows a transfer permission / prohibition flag setting screen. It is a figure which shows a table definition confirmation and edit screen.

  Hereinafter, an embodiment according to the present invention will be described with reference to FIGS. 1 to 9.

First, the configuration of a data migration system according to an embodiment of the present invention will be described with reference to FIGS.
In the data migration system 100, as shown in the configuration diagram of FIG. 1, a mainframe 101, a database server 103, a plurality of application server groups 105, and a configuration information management conversion server 106 are connected via a network 107. It is a configuration.

  Here, it is assumed that the mainframe 101 is running a structural database 102 (for example, XDM / SD (registered trademark)) on a mainframe OS (for example, VOS3 / US). Further, it is assumed that the database server 103 is running a relational database 104 (for example, HiRDB (registered trademark)) on an open system OS (for example, Linux). This will be explained in detail later.

  The configuration information management conversion server 106 is a server that provides a function of converting the structural database 102 on the mainframe 101 into a relational database 104 on the database server 103 equivalent thereto. For this purpose, the configuration information management conversion server 106 has an area for storing database configuration information of the structural database 102, configuration information obtained by converting the configuration information for the relational database 104, and other setting information. In addition, a program for acquiring database configuration information, a program for converting configuration information, and the like are installed. Examples of database configuration information include, for example, the data identification number, data length, and data type of each data on the database.

Next, the hardware / software configuration of the configuration information management conversion server will be described with reference to FIG.
As shown in FIG. 2, the configuration information management conversion server 106 can be realized by general computer hardware, and includes an arithmetic processing unit 201, a main memory 202, and a display I / F (Interface) unit 203. The input / output I / F unit 204, the communication I / F unit 205, and the storage unit 230 are connected via a communication path such as a BUS (bus). The arithmetic processing unit 201 is configured by a CPU (Central Processing Unit), refers to work data on the main memory 202, executes a loaded program, performs various determinations and arithmetic processing, and performs a configuration information management conversion server. Each part of 106 is controlled. The main memory 202 is composed of a volatile semiconductor storage device such as a RAM (Random Access Memory), and operates at a high speed but has a relatively small capacity, and stores work data and programs referred to by the arithmetic processing unit 201. The display I / F unit 203 is a part that is connected to an LCD (Liquid Crystal Display) display device 221 of the user terminal 220. The input / output I / F unit 204 is a part connected to an input / output device (for example, a keyboard 222 and a mouse 223) of the management terminal 220. The communication I / F unit 205 is a part that is connected to a network and handles an interface with another host device.

  The storage unit 230 includes a storage device using a magnetic storage medium such as an HDD (Hard Disk Drive) or a relatively large capacity storage device such as a nonvolatile semiconductor storage device such as an SSD. In the storage unit 230, a configuration information acquisition program 206, a table definition link information generation program 207, a database conversion program 208, and a database update extraction program 209 are installed as a program group, and mainframe database configuration information is used as a database. A table 210, a main frame record configuration information table 211, a table definition link information table 212, and an update table definition link information table 213 are stored.

  The configuration information acquisition program 206 is a program that extracts the configuration information of the structural database on the main frame 101. The table definition link information generation program 207 is a program that generates relation information between the structural type database 102 and the relational type database 104 equivalent thereto. The database conversion program 208 is a program for generating a relational type database 104 equivalent to the structural type database 102 based on the generated relational information. The database update extraction program 209 is a program for extracting an updated portion of the structural database 102 on the mainframe from a certain time point to a certain time point.

  Each table in the database will be described in detail later.

Next, the outline of the structural database and the relational database will be described with reference to FIG.
As shown in FIG. 3A, the structural database manages a basic unit of information existing in the database as a record. A record is a logical unit of information, and at the same time a unit for retrieving and updating data. Records have a parent-child relationship and are managed hierarchically. The record model is a record type Record consists of individual data items. The data item is the smallest unit of information that can be handled in the database. One data item has only one value, and each data type to be stored has an attribute called a data type.

  On the other hand, the relational database is a model database in which one piece of data is expressed as a set of a plurality of attribute values, and the data is stored by enumerating the sets. Usually, it is shown in the form of a table with attributes as columns and pairs as rows. A group of data that combines related attribute values is called a record, and the individual attribute values that make up a record are called fields. . When expressed in the form of a table, records correspond to rows and fields correspond to columns.

  For example, in the example of FIG. 3B, a database including a customer table having fields of customer ID, name, address, and telephone number and a sales table having fields of customer ID, product number, product, price, and sales date is shown. Has been. A relational database is characterized by being able to flexibly handle complex data and large-scale data by associating relationships between tables by a field having a unique value called a primary key. In this example, “customer ID” serves as a main key, and customer information and product information sold to the customer can be handled in association with each other.

  As described above, since the structured database and the relational database have semantic gaps, it is a problem to absorb this when the database is migrated.

Next, a data structure used in the data migration system will be described with reference to FIGS.
The mainframe database configuration information table 210 is a table for storing the database configuration information of the structural type database 102 acquired by the configuration information acquisition program 206. As shown in FIG. The fields of record identification No. 302, record name 303, record length 304, and migration enable / disable flag 305 are stored in association with each other.

  The data acquisition date 301 stores the data income date when the configuration information for each database for comparison with the latest database configuration information is acquired when the database configuration is changed. The record identification number 302 stores the identification number of each record in the structural database numbered by the configuration information acquisition program 205. The record name stores a name for identifying each record. The record length 304 stores the length of each record in bytes. The migration permission / prohibition flag 305 stores a flag (ON: migration, OFF: migration) indicating whether or not the corresponding record is to be linked to the relational database 104.

  The mainframe record configuration information table 211 is a table for storing information related to data items that are record configuration information of the structural database 102 acquired by the configuration information acquisition program 206, as shown in FIG. , Data acquisition date 401, record identification No. 402, data identification No. 403, data start position 404, data end position 405, data length 406, and data type 407 are stored in association with each other.

  The data acquisition date 401 stores the data income date when the configuration information for each database for comparison with the latest database configuration information when the database configuration is changed. The record identification number 402 stores the identification number of each record in the structural database numbered by the configuration information acquisition program 205. The data identification number 403 stores the identification number of each data item included in the corresponding record. In the data start position 404, the data end position 405, and the data length 406, the start position, end position, and data length from the record of the corresponding data item are stored in units of bytes. In the data type 407, the data type of the data item is stored as “integer”, “character string”, “hexadecimal data”, and so on.

  The table definition link information table 212 is a table for linking and storing the record definition of the structural type database 102, the table definition for the relational type database 104, and the data configuration information, as shown in FIG. , Record identification No. 501, data identification No. 502, table name 503, data name 504, data length 505, and data type 506 are stored in association with each other.

  In record identification No. 501 and data identification No. 502, the contents of record identification No. 402 and data identification No. 403 acquired from the mainframe record configuration information table 211 are stored. The table name 503 stores a name for identifying a table on the relational database 104. The data name 504 stores a name for identifying data of each field belonging to the table name 503. The data length 505 stores the contents of the data length acquired from the mainframe record configuration information table 211. The data type 506 stores a data type identifier obtained by converting the data type 407 of the mainframe record configuration information table 211 into a data type used for the relational database.

  The table definition and data configuration information for the relational database 104 are generated from the information stored in the mainframe database configuration information table 210 and the mainframe record configuration information storage unit 211 by the table definition link generation program 207, and this table definition It is stored in the link information table 212.

  As shown in FIG. 7, the update table definition link information table 213 has the same configuration as the table definition link information table 212. When the structural database 102 is changed, the update table definition link information table 213 is updated by the database update extraction program 209. Only the link information to be changed is stored. The configuration example of the update table definition link information table 213 is the same as that of the table definition link information table 212.

  Next, processing of the data migration system will be described with reference to FIGS.

First, processing when a database is transferred from a structural database to a relational database will be described with reference to FIGS. 8, 10, and 11.
In the process shown in FIG. 8, the data migration system of this embodiment is installed, and the definition information of the existing structural database 102 is changed to the newly constructed relational database 104 by the first data migration. This is a process of converting and reflecting the definition information in the relational database 104.

  First, the configuration information management conversion server 106 starts and executes the configuration information acquisition program 206 to store the database configuration information of the structural database 102 in the mainframe database configuration information table 210 shown in FIG. Then, the record configuration information of the structural database 102 is stored in the mainframe record configuration information table 211 shown in FIG. 5 (S701).

  Next, the configuration information management conversion server 106 displays the migration permission / prohibition flag setting screen 900 shown in FIG. 10 on the user terminal 220, and sets a flag as to whether or not to link the corresponding record to the relational database (S702). ). Records whose migration flag is set to ON are processed by the conversion program 206, the database conversion program 208, and the database update extraction program 209. If the flag indicating that the migration is allowed is set to OFF, the program is not processed.

  In the migration permission / prohibition flag setting screen 900, as shown in FIG. 10, the record identification No. 901 and the record name 902 of each record acquired from the mainframe database configuration information table 210 are displayed, the user confirms the corresponding record, and the corresponding record It is possible to input ON / OFF of the migration propriety flag 903. When the user performs an operation of pressing the setting button 904 with the mouse 223 or the like, the information of the migration possibility flag 903 input by the user is stored in the migration possibility flag 305 of the mainframe database configuration information table 210.

  Next, the configuration information management conversion server 106 starts and executes the table definition link information generation program 207 to convert the database structure of the structural database 102 and reflect the definition information to the relational database 104. Information is stored in the table definition link information table 212 (S703). The table definition link information generation program 207 creates a table definition and a data definition for the relational type database 104 from each record definition of the structural type database 102 and each data definition belonging to the record definition, and links them with each other. 6 is stored in the table definition link information table 212 shown in FIG.

  Next, the configuration information management conversion server 106 displays the table definition confirmation / edit screen 1000 on the user terminal 202, confirms the table name and field name to the user, and changes the user if necessary (S704). The table name set by the table definition link information generation program 207 is the identification number of the structural database or the name automatically set by the program. In this way, it is possible to set a name that can be easily identified by the user. As shown in FIG. 11, the table definition confirmation / edit screen 1000 is used to edit a character string of record identification No. 1001 and record name 1002 that are record identification information of the structural database 102 stored in the table definition link information table 212. The table name 1003 and field name 1004 linked to it are displayed in a state where the character string can be edited. When the user confirms the table name 1003 and the field name 1004, changes it to an arbitrary character string according to the database naming rules, and presses the enter button 1005 with the mouse 223 or the like, the table definition link information table 212 is displayed. The corresponding data of is updated.

  Then, the configuration information management conversion server 106 starts and executes the database conversion program 208, and thereby the relational database 104 equivalent to the database of the structural database 102 based on the information stored in the table definition link information table 212. The database table is created (S705). Further, the database conversion program 208 may create a script and execute the script, so that the database migration from the structural database 102 to the relational database 104 may be performed.

Next, the processing when the structured database is updated or the settings related to migration are changed will be described with reference to FIG.
The flowchart shown in FIG. 9 shows the case where the structural database is changed after the initial database migration from the structural database 102 to the relational database 104, or the migration permission flag 305 of the mainframe database configuration information table 210 is changed. It shows processing executed when the flag is changed from OFF (not shifted) to ON (transferred).

  First, when the structural database 102 is changed, the configuration information management conversion server 106 activates and executes the configuration re-extraction program 209, whereby the mainframe database configuration information table 210 and the mainframe record configuration information table 211 are obtained. The updated location is extracted (S801).

  Further, when the flag of the migration permission flag 405 of the mainframe database configuration information table 210 is changed from OFF to ON, the necessary parts of the mainframe database configuration information table 210 and the mainframe record configuration information table 211 are changed as in S801. Extract (S802).

  Next, the configuration information management conversion server 106 starts and executes the table definition link information generation program 207, thereby converting the database structure of the location extracted in S801 and S802, and transferring the definition information to the relational database 104. Information to be reflected is stored in the update table definition link information table 213 shown in FIG. 7 (S803).

  Next, the configuration information management conversion server 106 displays the table definition confirmation / edit screen 1000 on the user terminal 202, confirms the table name and field name to the user, and changes the user if necessary (S804). This is the same as S704.

  Then, the configuration information management conversion server 106 starts up and executes the database conversion program 208, and based on the information stored in the update table definition link information table 213, the relational type equivalent to the database of the structural type database 102 A database table of the database 104 is created (S705). This is the same as S705.

  DESCRIPTION OF SYMBOLS 100 ... Data migration system, 101 ... Main frame, 102 ... Structural type database, 103 ... Database server, 104 ... Relational type database, 105 ... Application server group, 106 ... Configuration information management conversion server, 107 ... Network

Claims (5)

Means for reading the data structure and data of a structured database;
A data migration system comprising a data structure of a structured database and a means for converting the data structure and data of a relational database equivalent thereto into data.   2. The data migration system according to claim 1, further comprising means for associating and holding records and data items of the structural database and tables and fields of the relational database.   And a means for obtaining an update of the structural database from a certain point in time and converting the data structure and data of the updated structural database into a data structure and data of a relational database equivalent thereto. 1. The data migration system according to 1.   2. The data migration system according to claim 1, further comprising means for designating whether or not to make a data migration target for a record in a structured database, wherein the record to be migrated is a migration target. Means for displaying the table name and field name of the relational database to be migrated;
The data migration system according to claim 2, further comprising means for changing the table name and the field name.

Images